How to prevent sporadic ringing on VLXSMPS pin in "Direct SMPS supply" configuration?

[Manuel Del Basso]

Hello,

we use the system supply configuration "Direct SMPS supply" (AN5419 Figure 2) for the STM32H735VG.

We have the following problem in the current hardware design:

At pin VLXSMPS an oscillation with approx. 6.4MHz can be measured sporadically (see DS1Z_QuickPrint1) which we cannot get a grip on. The ringing does not occur after each switching cycle (see DS1Z_QuickPrint2) so this is meant sporadically.

The following component is used as the inductor L900:

2.2uH/1.8A/116mR

DFE201612E-2R2M=P2

Mouser part number: 81-DFE201612E-2R2MP2

The following component from Murata is used as capacitors C913 and C914:

10µF 0402 Low ESR (0.006Ohm@3MHz)

GRM155R60J106ME47

ESR diagram: https://www.murata.com/en-eu/products/productdetail?partno=GRM155R60J106ME47%23

The 220pF on the VLXSMPS pin is populated.

What can we do to suppress the oscillation?

Sincerely

Manuel Del Basso

[Peter BENSCH]

Such oscillations are a typical behaviour when switching off buck regulators, where the SMPS inductance and the capacitances (internal switching capacitances, external capacitors) form an oscillating circuit.

In the context of the STM32, as also mentioned in AN5419 table 2, the current path from VLXSMPS to VFBSMPS and from their capacitors to GND should be made as short and wide as possible in the layout and without vias in order to reduce such oscillations as best as possible.

Regards

/Peter

[Manuel Del Basso]

Thank you for your quick response.

So far I only know the ringing in a constant form, so that after each switching off of the SMPS controller output stage a ringing is measurable.

Is it actually normal behavior that the ringing only occurs with some switching cycles?

L900 is currently placed on the bottom layer, the STM32H7 is on the top layer. A conversion of the power supply might be conceivable, thanks for the tip.

sincerly Manuel

[Peter BENSCH · Accepted solution]

When you see the square waveform, the included synchronous switching regulator switches between Hi and Lo. As soon as the ringing occurs, it is in tristate mode, presumably because currently no additionally energy needs to be supplied due to lower load.

Regards

/Peter

[Manuel Del Basso]

Thanks for the hint. It is simply a buck converter in discontinuous operation.

sincerly manual

[Peter BENSCH]

Absolutely correct. :smiling_face_with_smiling_eyes:

If the problem is solved, please mark this thread as answered by selecting Select as best, as also explained here. This will help other users find that answer faster.

Regards

/Peter